Comparison of Accelerometer Cut Points for Predicting Activity Intensity in Youth

Department of Nutrition and Exercise Sciences, Oregon State University, Corvallis, OR 97331, USA.
Medicine and science in sports and exercise (Impact Factor: 3.98). 12/2010; 43(7):1360-8. DOI: 10.1249/MSS.0b013e318206476e
Source: PubMed


The absence of comparative validity studies has prevented researchers from reaching consensus regarding the application of intensity-related accelerometer cut points for children and adolescents.
This study aimed to evaluate the classification accuracy of five sets of independently developed ActiGraph cut points using energy expenditure, measured by indirect calorimetry, as a criterion reference standard.
A total of 206 participants between the ages of 5 and 15 yr completed 12 standardized activity trials. Trials consisted of sedentary activities (lying down, writing, computer game), lifestyle activities (sweeping, laundry, throw and catch, aerobics, basketball), and ambulatory activities (comfortable walk, brisk walk, brisk treadmill walk, running). During each trial, participants wore an ActiGraph GT1M, and V˙O2 was measured breath-by-breath using the Oxycon Mobile portable metabolic system. Physical activity intensity was estimated using five independently developed cut points: Freedson/Trost (FT), Puyau (PU), Treuth (TR), Mattocks (MT), and Evenson (EV). Classification accuracy was evaluated via weighted κ statistics and area under the receiver operating characteristic curve (ROC-AUC).
Across all four intensity levels, the EV (κ=0.68) and FT (κ=0.66) cut points exhibited significantly better agreement than TR (κ=0.62), MT (κ=0.54), and PU (κ=0.36). The EV and FT cut points exhibited significantly better classification accuracy for moderate- to vigorous-intensity physical activity (ROC-AUC=0.90) than TR, PU, or MT cut points (ROC-AUC=0.77-0.85). Only the EV cut points provided acceptable classification accuracy for all four levels of physical activity intensity and performed well among children of all ages. The widely applied sedentary cut point of 100 counts per minute exhibited excellent classification accuracy (ROC-AUC=0.90).
On the basis of these findings, we recommend that researchers use the EV ActiGraph cut points to estimate time spent in sedentary, light-, moderate-, and vigorous-intensity activity in children and adolescents.

Download full-text


Available from: Karin Pfeiffer, Feb 14, 2015
  • Source
    • "Accelerometer cutpoints for sedentary activity are flat with values of 100 counts per minute from the Actigraph and are now being widely adopted and accepted (Fischer et al. 2012; Trost et al. 2011). Trost and colleagues showed how a flat value of 100 counts per minute was associated with an AUC of 0.90 when compared with measured energy expenditure of pre-defined sedentary activities (e.g., writing while sitting) (Trost et al. 2011). In another study, Ridgers et al. (2012) found 96 counts per minute to be the optimal cutpoint to determine sitting time when compared to classifications obtained from the activPAL (AUC = 0.75). "
    [Show abstract] [Hide abstract]
    ABSTRACT: The study compares MET-defined cutpoints used to classify sedentary behaviors in children using a simulated free-living design. A sample of 102 children (54 boys and 48 girls; 7-13 years) completed a set of 12 activities (randomly selected from a pool of 24 activities) in a random order. Activities were predetermined and ranged from sedentary to vigorous intensities. Participant's energy expenditure was measured using a portable indirect calorimetry system, Oxycon mobile. Measured minute-by-minute VO2 values (i.e., ml/kg/min) were converted to an adult- or child-MET value using the standard 3.5 ml/kg/min or the estimated child resting metabolic rate, respectively. Classification agreement was examined for both the "standard" (1.5 adult-METs) and an "adjusted" (2.0 adult-METs) MET-derived threshold for classifying sedentary behavior. Alternatively, we also tested the classification accuracy of a 1.5 child-MET threshold. Classification accuracy of sedentary activities was evaluated relative to the predetermined intensity categorization using receiver operator characteristic curves. There were clear improvements in the classification accuracy for sedentary activities when a threshold of 2.0 adult-METs was used instead of 1.5 METs (Se1.5 METs = 4.7 %, Sp1.5 METs = 100.0 %; Se2.0 METs = 36.9 %, Sp2.0 METs = 100.0 %). The use of child-METs while maintaining the 1.5 threshold also resulted in improvements in classification (Se = 45.1 %, Sp = 100.0 %). Adult-MET thresholds are not appropriate for children when classifying sedentary activities. Classification accuracy for identifying sedentary activities was improved when either an adult-MET of 2.0 or a child-MET of 1.5 was used.
    Arbeitsphysiologie 08/2015; DOI:10.1007/s00421-015-3238-1 · 2.19 Impact Factor
  • Source
    • "At the time of the study, the baseline accelerometry data was analysed using the Puyau cut points (as they were the most applicable to our age group) and fed back to parents as part of the intervention protocol. Thus, this data was analysed prior to the recommendations published by Trost et al in 2011. We chose to keep the cut points consistent across methods and thus used the Puyau cut points for the other methods presented in this paper. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Although accelerometers can assess sleep and activity over 24 h, sleep data must be removed before physical activity and sedentary time can be examined appropriately. We compared the effect of 6 different sleep-scoring rules on physical activity and sedentary time. Activity and sleep were obtained by accelerometry (ActiGraph GT3X) over 7 days in 291 children (51.3% overweight or obese) aged 4-8.9 years. Three methods removed sleep using individualised time filters and two methods applied standard time filters to remove sleep each day (9 pm-6 am, 12 am-6 am). The final method did not remove sleep but simply defined non-wear as at least 60 min of consecutive zeros over the 24-h period. Different methods of removing sleep from 24-h data markedly affect estimates of sedentary time, yielding values ranging from 556 to 1145 min/day. Estimates of non-wear time (33-193 min), wear time (736-1337 min) and counts per minute (384-658) also showed considerable variation. By contrast, estimates of moderate-to-vigorous activity (MVPA) were similar, varying by less than 1 min/day. Different scoring methods to remove sleep from 24-h accelerometry data do not affect measures of MVPA, whereas estimates of counts per minute and sedentary time depend considerably on which technique is used.
    Journal of Sports Sciences 07/2015; DOI:10.1080/02640414.2015.1068438 · 2.25 Impact Factor
  • Source
    • "CPM = counts per minute; SED = sedentary time; LPA = light physical activity; MVPA = moderate-to-vigorous physical activity; LVPA = light-to-vigorous physical activity; ICC = intraclass correlation; LoA = 95% limits of agreement. The study was conducted in Sogn og Fjordane, Norway, 2014. in external validation studies in youth (5–15 years of age) (Trost et al., 2011) and preschool (4–6 years of age) (Janssen et al., 2013) samples. Janssen et al. (2013) also found the Pate et al. (2006) MVPA cut point (≥1680 cpm) developed in preschool children to perform well, however, applying this cut point to our data did not change any findings in terms of reliability. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective: To determine the intra-individual agreement for objectively measured physical activity (PA) and sedentary behavior (SED) over two subsequent weeks in preschool children. Method: Ninety-one children aged 3 to 5 years (49% boys) from three preschools in Sogn og Fjordane, Norway, provided 14 consecutive days of accelerometer data (Actigraph GT3X +) during the autumn of 2014. Week-by-week reliability was assessed using intraclass correlation (ICC), Bland-Altman plots and 95% limits of agreement for different wear time criteria (≥ 6, 8 and 10 h/day and ≥ 3 and 5 days/week). Results: The week-by-week ICC was ≥ 0.75 for all variables across all wear criteria applied, except for absolute sedentary time (ICC 0.61-0.81). Using a ≥ 8 h/day and ≥ 3 days/week criterion (n = 78), limits of agreement were ± 209.5 cpm for overall PA, ± 68.6 min/day for SED, ± 43.8 min/day for light PA, ± 20.2 min/day for moderate-to-vigorous PA, and ± 55.9 min/day for light-to-vigorous PA, equaling 1.0-1.6 standard deviation units. Conclusion: Considerable week-by-week variability was found for all variables. Researchers need to be aware of substantial intra-individual variability in accelerometer-measurements and take necessary actions according to the hypothesis under study, as noise in any measurement will preclude researchers' ability to arrive at valid conclusions in epidemiology.
Show more